Do We Train on Test Data? The Impact of Near-Duplicates on License Plate Recognition

Autor:	Laroca, Rayson, Estevam, Valter, Britto Jr., Alceu S., Minetto, Rodrigo, Menotti, David
Rok vydání:	2023
Předmět:	Computer Science - Computer Vision and Pattern Recognition
Druh dokumentu:	Working Paper
DOI:	10.1109/IJCNN54540.2023.10191584
Popis:	This work draws attention to the large fraction of near-duplicates in the training and test sets of datasets widely adopted in License Plate Recognition (LPR) research. These duplicates refer to images that, although different, show the same license plate. Our experiments, conducted on the two most popular datasets in the field, show a substantial decrease in recognition rate when six well-known models are trained and tested under fair splits, that is, in the absence of duplicates in the training and test sets. Moreover, in one of the datasets, the ranking of models changed considerably when they were trained and tested under duplicate-free splits. These findings suggest that such duplicates have significantly biased the evaluation and development of deep learning-based models for LPR. The list of near-duplicates we have found and proposals for fair splits are publicly available for further research at https://raysonlaroca.github.io/supp/lpr-train-on-test/ Comment: Accepted for presentation at the International Joint Conference on Neural Networks (IJCNN) 2023
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2304.04653 Zobrazit plný text záznamu View this record from Arxiv